palimpzest 1.2.0__py3-none-any.whl → 1.3.0__py3-none-any.whl

palimpzest/constants.py

@@ -31,9 +31,9 @@ class Model(str, Enum):
     GEMINI_2_0_FLASH = "vertex_ai/gemini-2.0-flash"
     GEMINI_2_5_FLASH = "vertex_ai/gemini-2.5-flash"
     GEMINI_2_5_PRO = "vertex_ai/gemini-2.5-pro"
-    GOOGLE_GEMINI_2_5_FLASH = "google/gemini-2.5-flash"
-    GOOGLE_GEMINI_2_5_FLASH_LITE = "google/gemini-2.5-flash-lite"
-    GOOGLE_GEMINI_2_5_PRO = "google/gemini-2.5-pro"
+    GOOGLE_GEMINI_2_5_FLASH = "gemini/gemini-2.5-flash"
+    GOOGLE_GEMINI_2_5_FLASH_LITE = "gemini/gemini-2.5-flash-lite"
+    GOOGLE_GEMINI_2_5_PRO = "gemini/gemini-2.5-pro"
     LLAMA_4_MAVERICK = "vertex_ai/meta/llama-4-maverick-17b-128e-instruct-maas"
     GPT_4o_AUDIO_PREVIEW = "openai/gpt-4o-audio-preview"
     GPT_4o_MINI_AUDIO_PREVIEW = "openai/gpt-4o-mini-audio-preview"
@@ -72,8 +72,8 @@ class Model(str, Enum):
     def is_vertex_model(self):
         return "vertex_ai" in self.value.lower()
 
-    def is_google_model(self):
-        return "google" in self.value.lower()
+    def is_google_ai_studio_model(self):
+        return "gemini/" in self.value.lower()
 
     def is_vllm_model(self):
         return "hosted_vllm" in self.value.lower()

palimpzest/query/processor/config.py

@@ -27,7 +27,7 @@ class QueryProcessorConfig(BaseModel):
     join_parallelism: int = Field(default=64)
     batch_size: int | None = Field(default=None)
     reasoning_effort: str | None = Field(default=None)  # Gemini: "disable", "low", "medium", "high"
-    use_vertex: bool = Field(default=True)  # Whether to use Vertex models for Gemini or Google models
+    use_vertex: bool = Field(default=False)  # Whether to use Vertex models for Gemini or Google models
     gemini_credentials_path: str | None = Field(default=None)  # Path to Gemini credentials file
     api_base: str | None = Field(default=None)  # API base URL for vLLM
 

palimpzest/query/processor/query_processor_factory.py

@@ -2,6 +2,8 @@ import logging
 import os
 from enum import Enum
 
+from dotenv import load_dotenv
+
 from palimpzest.core.data.dataset import Dataset
 from palimpzest.core.elements.records import DataRecordCollection
 from palimpzest.query.execution.execution_strategy import ExecutionStrategy, SentinelExecutionStrategy
@@ -108,7 +110,7 @@ class QueryProcessorFactory:
                 raise ValueError("ANTHROPIC_API_KEY must be set to use Anthropic models.")
             if model.is_together_model() and not together_key:
                 raise ValueError("TOGETHER_API_KEY must be set to use Together models.")
-            if model.is_google_model() and not (gemini_key or google_key or config.gemini_credentials_path):
+            if model.is_google_ai_studio_model() and not (gemini_key or google_key or config.gemini_credentials_path):
                 raise ValueError("GEMINI_API_KEY, GOOGLE_API_KEY, or gemini_credentials path must be set to use Google Gemini models.")
             if model.is_vllm_model() and config.api_base is None:
                 raise ValueError("api_base must be set to use vLLM models.")
@@ -194,6 +196,7 @@ class QueryProcessorFactory:
         train_dataset: dict[str, Dataset] | None = None,
         validator: Validator | None = None,
     ) -> DataRecordCollection:
+        load_dotenv(override=True)
         logger.info(f"Creating processor for dataset: {dataset}")
         processor = cls.create_processor(dataset, config, train_dataset, validator)
         logger.info(f"Created processor: {processor}")

palimpzest/utils/model_helpers.py

@@ -3,13 +3,12 @@ import os
 from palimpzest.constants import Model
 
 
-# TODO: better handle vertex vs. google for gemini models
-def get_models(include_embedding: bool = False, use_vertex: bool = True, gemini_credentials_path: str | None = None, api_base: str | None = None) -> list[Model]:
+def get_models(include_embedding: bool = False, use_vertex: bool = False, gemini_credentials_path: str | None = None, api_base: str | None = None) -> list[Model]:
     """
     Return the set of models which the system has access to based on the set environment variables.
     """
     models = []
-    if os.getenv("OPENAI_API_KEY") is not None:
+    if os.getenv("OPENAI_API_KEY") not in [None, ""]:
         openai_models = [model for model in Model if model.is_openai_model()]
         if not include_embedding:
             openai_models = [
@@ -17,7 +16,7 @@ def get_models(include_embedding: bool = False, use_vertex: bool = True, gemini_
             ]
         models.extend(openai_models)
 
-    if os.getenv("TOGETHER_API_KEY") is not None:
+    if os.getenv("TOGETHER_API_KEY") not in [None, ""]:
         together_models = [model for model in Model if model.is_together_model()]
         if not include_embedding:
             together_models = [
@@ -25,7 +24,7 @@ def get_models(include_embedding: bool = False, use_vertex: bool = True, gemini_
             ]
         models.extend(together_models)
 
-    if os.getenv("ANTHROPIC_API_KEY") is not None:
+    if os.getenv("ANTHROPIC_API_KEY") not in [None, ""]:
         anthropic_models = [model for model in Model if model.is_anthropic_model()]
         if not include_embedding:
             anthropic_models = [
@@ -38,9 +37,9 @@ def get_models(include_embedding: bool = False, use_vertex: bool = True, gemini_
         if gemini_credentials_path is None
         else gemini_credentials_path
     )
-    if os.getenv("GEMINI_API_KEY") is not None or os.path.exists(gemini_credentials_path):
+    if os.getenv("GEMINI_API_KEY") not in [None, ""] or (use_vertex and os.path.exists(gemini_credentials_path)):
         vertex_models = [model for model in Model if model.is_vertex_model()]
-        google_models = [model for model in Model if model.is_google_model()]
+        google_ai_studio_models = [model for model in Model if model.is_google_ai_studio_model()]
         if not include_embedding:
             vertex_models = [
                 model for model in vertex_models if not model.is_embedding_model()
@@ -48,7 +47,7 @@ def get_models(include_embedding: bool = False, use_vertex: bool = True, gemini_
         if use_vertex:
             models.extend(vertex_models)
         else:
-            models.extend(google_models)
+            models.extend(google_ai_studio_models)
 
     if api_base is not None:
         vllm_models = [model for model in Model if model.is_vllm_model()]

{palimpzest-1.2.0.dist-info → palimpzest-1.3.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: palimpzest
-Version: 1.2.0
+Version: 1.3.0
 Summary: Palimpzest is a system which enables anyone to process AI-powered analytical queries simply by defining them in a declarative language
 Author-email: MIT DSG Semantic Management Lab <michjc@csail.mit.edu>
 Project-URL: homepage, https://palimpzest.org
@@ -34,6 +34,7 @@ Requires-Dist: PyLD>=2.0.4
 Requires-Dist: pyarrow>=20.0.0
 Requires-Dist: pypdf>=5.1.0
 Requires-Dist: pytest-mock>=3.14.0
+Requires-Dist: python-dotenv>=1.2.1
 Requires-Dist: pyyaml>=6.0.1
 Requires-Dist: requests>=2.25
 Requires-Dist: ruff>=0.9.0

{palimpzest-1.2.0.dist-info → palimpzest-1.3.0.dist-info}/RECORD

@@ -1,5 +1,5 @@
 palimpzest/__init__.py,sha256=1PzadDDOVMQJKNEYUH0_tw8tQKUYTT31M0vuzTr2Rqk,1694
-palimpzest/constants.py,sha256=R_vNt3FW9MA4-HuszrGyGtDLRE9YMLXRMwX5GZcRH38,23323
+palimpzest/constants.py,sha256=kfuDTdSdeTCWp2JJopeuwpGIf8MC6jQ8SDBFuRIhN3k,23334
 palimpzest/policy.py,sha256=lIvw_C_rmwCH4LZaeNkAuixl8zw9RAW_JcSWSHPjKyc,11628
 palimpzest/agents/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 palimpzest/agents/compute_agents.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -71,9 +71,9 @@ palimpzest/query/optimizer/primitives.py,sha256=jMMVq37y1tWiPU1lSSKQP9OP-mzkpSxS
 palimpzest/query/optimizer/rules.py,sha256=awhe76trskv5Tq5E2QHpUN_YV6jH8INywa0Ige8IIhY,53341
 palimpzest/query/optimizer/tasks.py,sha256=DNJjY2QldfKFWj6INHElMh88dYc36Z5m3wHwbs4jyF4,30455
 palimpzest/query/processor/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-palimpzest/query/processor/config.py,sha256=8-MpPYHv2SI4dub4MP_gOYSRxO80_ALLuWRxD-F2YOg,2521
+palimpzest/query/processor/config.py,sha256=rVpXNfzJvYstNZ2PxhQMWBpHCmjLDfiKh4ERKWRojc0,2522
 palimpzest/query/processor/query_processor.py,sha256=T4ffPbnOX23G8FDITzmM7Iw7DUEDWIHnwl8XLYllgjg,6240
-palimpzest/query/processor/query_processor_factory.py,sha256=l9f0C0lngOihZDzH0TK9WdKR9CwwgB6IbNZftonSFR0,9576
+palimpzest/query/processor/query_processor_factory.py,sha256=qED6pJtJJXVci2a4nScURSpJohrz1LIW6tXY05un2R8,9653
 palimpzest/schemabuilder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 palimpzest/schemabuilder/schema_builder.py,sha256=QraGp66dcD-ej6Y2mER40o86G9JqlBkL7swkJzjUAIY,7968
 palimpzest/tools/README.md,sha256=56_6LPG80uc0CLVhTBP6I1wgIffNv9cyTr0TmVZqmrM,483
@@ -84,13 +84,13 @@ palimpzest/tools/skema_tools.py,sha256=HXUFpjMhbVxZwKKkATeK-FwtlTCawaCbeP-uHntI1
 palimpzest/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 palimpzest/utils/env_helpers.py,sha256=n81KzoJ459pRxo7QmJA7duazwWsfoMGTHc71D2LatFk,334
 palimpzest/utils/hash_helpers.py,sha256=3A8dA7SbXTwnnvZvPVNqqMLlVRhCKyKF_bjNNAu3Exk,334
-palimpzest/utils/model_helpers.py,sha256=X6SlMgD5I5Aj_cxaFaoGaaNvOOqTNZVmjj6zbfn63Yk,2476
+palimpzest/utils/model_helpers.py,sha256=SqcY8rWzZ7D3Vgeq8d4OGNLGv4dXXVMWiJDqSCaalRQ,2490
 palimpzest/utils/progress.py,sha256=eHXrTPTCRHjMdK0EjYRUzSxcV6N1lK8TS3Ju_ZlQLhY,22002
 palimpzest/utils/udfs.py,sha256=LjHic54B1az-rKgNLur0wOpaz2ko_UodjLEJrazkxvY,1854
 palimpzest/validator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 palimpzest/validator/validator.py,sha256=SvjK09zCpGtK0yM0OasvQlSzyq3loy32DyOOKRmYXC0,15977
-palimpzest-1.2.0.dist-info/licenses/LICENSE,sha256=5GUlHy9lr-Py9kvV38FF1m3yy3NqM18fefuE9wkWumo,1079
-palimpzest-1.2.0.dist-info/METADATA,sha256=IKxg8RllEvn6dgboEJVnxdnd5RwYmXFhIL2FHvoYpWw,5359
-palimpzest-1.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-palimpzest-1.2.0.dist-info/top_level.txt,sha256=raV06dJUgohefUn3ZyJS2uqp_Y76EOLA9Y2e_fxt8Ew,11
-palimpzest-1.2.0.dist-info/RECORD,,
+palimpzest-1.3.0.dist-info/licenses/LICENSE,sha256=5GUlHy9lr-Py9kvV38FF1m3yy3NqM18fefuE9wkWumo,1079
+palimpzest-1.3.0.dist-info/METADATA,sha256=1VcCZTtJHg1ujb41-3QquXQLYaWtpCG9AHuvtrvJGvg,5395
+palimpzest-1.3.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+palimpzest-1.3.0.dist-info/top_level.txt,sha256=raV06dJUgohefUn3ZyJS2uqp_Y76EOLA9Y2e_fxt8Ew,11
+palimpzest-1.3.0.dist-info/RECORD,,